PS295 Chapter Ten Summary
Analyzing Experimental Data
An Intuitive Approach to Analysis
If the IV has an effect on the DV we should find that the means for the experimental conditions
Different group averages would suggest that the IV had an effect; it created differences in the
behaviour in various conditions and resulted in systematic variance
Error Variance Can Cause Mean Differences
Problem: means of experimental conditions may differ even if the IV does NOT have an effect
Error variance reflects the random influences of variables that remain unidentified in the study
These uncontrolled and unidentified variables lead participants to respond differently whether
or not the IV has an effect.
If we can estimate how much the means of the conditions are expected to differ even if the IV
has no effect, then we can determine whether the difference we observe between the means
exceeds this estimate.
We can conclude that the IV has an effect when the difference between the means of the
experimental conditions is larger than we expect it to be when that difference is due solely to
the effects of error variance.
The Null Hypothesis
Researchers use inferential statistics to determine whether observed differences between the
means of the experimental conditions are greater than expected on the basis of error variance
The null hypothesis states that the IV did not have an effect on the DV and the opposite is
experimental hypothesis which states that the IV did have an effect
Rejecting the null hypothesis means that the researcher will conclude that the IV did have an
effect and failing to reject the null… means that the researcher will conclude that the IV has no
The researcher will fail to reject the null if analyses show a high probability that the difference
between the groups means reflects nothing more than the influence of error variance Type I and Type II Errors
Researcher may make decisions about the null and outcomes may result into four possible
(1) The researcher may correctly reject the null thereby showing a true effect of the IV
(2) the research may correctly fail to reject the null concluding that the IV has no effect
..the other two are type I and type II errors..
A type I error occurs when a researcher concludes that the IV has an effect on the DV when in
fact the observed difference between the means is actually due to error variance